SentenceTransformer based on BAAI/bge-base-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("datasocietyco/bge-base-en-v1.5-course-recommender-v1")
# Run inference
sentences = [
    'Clustering in NLP',
    'This course covers the clustering concepts of natural language processing, equipping learners with the ability to cluster text data into groups and topics by finding similarities between different documents.',
    'Course language: Python',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

Unnamed Dataset

Size: 183 training samples
Columns: name, description, languages, prerequisites, target_audience, and merged

Approximate statistics based on the first 183 samples:

	name	description	languages	prerequisites	target_audience	merged
type	string	string	string	string	string	string
details	min: 3 tokens mean: 7.06 tokens max: 16 tokens	min: 13 tokens mean: 40.5 tokens max: 117 tokens	min: 6 tokens mean: 6.66 tokens max: 10 tokens	min: 8 tokens mean: 12.56 tokens max: 21 tokens	min: 5 tokens mean: 23.2 tokens max: 54 tokens	min: 45 tokens mean: 81.98 tokens max: 174 tokens

Samples:

name	description	languages	prerequisites	target_audience	merged
`Foundations of Big Data`	`A theoretical course covering topics on how to handle data at scale and the different tools needed for distributed data storage, analysis, and management. Learners will be able to dive into the vast world of data and computing at scale and get a comprehensive overview of distributed computing.`	`Course language: TBD`	`Prerequisite course required: Optimizing Ensemble Methods`	`Professionals who would like to learn the core concepts of big data and understand data at scale`	`Foundations of Big Data A theoretical course covering topics on how to handle data at scale and the different tools needed for distributed data storage, analysis, and management. Learners will be able to dive into the vast world of data and computing at scale and get a comprehensive overview of distributed computing. Course language: TBD Prerequisite course required: Optimizing Ensemble Methods Professionals who would like to learn the core concepts of big data and understand data at scale`
`Big Data Orchestration & Workflow Management`	`A theoretical course covering topics on how to handle data at scale and the different tools needed for orchestrating big data systems and manage the workflow. Learners will be able to dive into the vast world of data and computing at scale and get a comprehensive overview of the distributed resource management ecosystem.`	`Course language: TBD`	`Prerequisite course required: Foundations of Big Data`	`Professionals who would like to learn the core concepts of distributed system orchestration and workflow management tools.`	Big Data Orchestration & Workflow Management A theoretical course covering topics on how to handle data at scale and the different tools needed for orchestrating big data systems and manage the workflow. Learners will be able to dive into the vast world of data and computing at scale and get a comprehensive overview of the distributed resource management ecosystem. Course language: TBD Prerequisite course required: Foundations of Big Data Professionals who would like to learn the core concepts of distributed system orchestration and workflow management tools.
`Distributed Data Storage (Hadoop)`	`A course that covers theory and implementation on a specific cloud platform covering topics on distributed data storage systems. Learners will be able to dive into the nature of storing and processing data at scale using tools like Hadoop on a selected cloud platform. This course will allow students to get a great foundation for creating and managing distributed data storage resources.`	`Course language: Java, Python`	`Prerequisite course required: Foundations of Big Data`	`Professionals who have coding knowledge and want to learn to create a scalable data storage solution using cloud services.`	Distributed Data Storage (Hadoop) A course that covers theory and implementation on a specific cloud platform covering topics on distributed data storage systems. Learners will be able to dive into the nature of storing and processing data at scale using tools like Hadoop on a selected cloud platform. This course will allow students to get a great foundation for creating and managing distributed data storage resources. Course language: Java, Python Prerequisite course required: Foundations of Big Data Professionals who have coding knowledge and want to learn to create a scalable data storage solution using cloud services.

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Evaluation Dataset

Unnamed Dataset

Size: 50 evaluation samples
Columns: name, description, languages, prerequisites, target_audience, and merged

Approximate statistics based on the first 50 samples:

	name	description	languages	prerequisites	target_audience	merged
type	string	string	string	string	string	string
details	min: 3 tokens mean: 6.98 tokens max: 15 tokens	min: 16 tokens mean: 39.66 tokens max: 83 tokens	min: 6 tokens mean: 6.66 tokens max: 10 tokens	min: 8 tokens mean: 12.58 tokens max: 21 tokens	min: 5 tokens mean: 24.06 tokens max: 54 tokens	min: 47 tokens mean: 81.94 tokens max: 139 tokens

Samples:

name	description	languages	prerequisites	target_audience	merged
`Word Embeddings in NLP`	`This course covers the intermediate concepts of natural language processing like creating word embeddings, feature engineering and word embeddings for finding text features for model development.`	`Course language: Python`	`Prerequisite course required: Topic Modeling in NLP`	`This is an intermediate level course for data scientists who have experience in NLP and want to learn to process and mine natural language and text data.`	`Word Embeddings in NLP This course covers the intermediate concepts of natural language processing like creating word embeddings, feature engineering and word embeddings for finding text features for model development. Course language: Python Prerequisite course required: Topic Modeling in NLP This is an intermediate level course for data scientists who have experience in NLP and want to learn to process and mine natural language and text data.`
`Big Data Orchestration & Workflow Management`	`A theoretical course covering topics on how to handle data at scale and the different tools needed for orchestrating big data systems and manage the workflow. Learners will be able to dive into the vast world of data and computing at scale and get a comprehensive overview of the distributed resource management ecosystem.`	`Course language: TBD`	`Prerequisite course required: Foundations of Big Data`	`Professionals who would like to learn the core concepts of distributed system orchestration and workflow management tools.`	Big Data Orchestration & Workflow Management A theoretical course covering topics on how to handle data at scale and the different tools needed for orchestrating big data systems and manage the workflow. Learners will be able to dive into the vast world of data and computing at scale and get a comprehensive overview of the distributed resource management ecosystem. Course language: TBD Prerequisite course required: Foundations of Big Data Professionals who would like to learn the core concepts of distributed system orchestration and workflow management tools.
`Accelerating Data Engineering Pipelines`	`Explore how to employ advanced data engineering tools and techniques with GPUs to significantly improve data engineering pipelines`	`Course language: Python`	`No prerequisite course required`	`Professionals who wants to learn the foundation of data science and lays the groundwork for analysis and modeling.`	`Accelerating Data Engineering Pipelines Explore how to employ advanced data engineering tools and techniques with GPUs to significantly improve data engineering pipelines Course language: Python No prerequisite course required Professionals who wants to learn the foundation of data science and lays the groundwork for analysis and modeling.`

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 3e-06
max_steps: 64
warmup_ratio: 0.1
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 3e-06
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 3.0
max_steps: 64
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss
1.6667	20	1.4345	1.0243
3.3333	40	0.9835	0.7613
5.0	60	0.7294	0.6593

Framework Versions

Python: 3.9.13
Sentence Transformers: 3.1.1
Transformers: 4.45.1
PyTorch: 2.2.2
Accelerate: 0.34.2
Datasets: 3.0.0
Tokenizers: 0.20.0

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}